Transducer Position Control System for Disk Storage Equipment

This invention relates to electronic positioning devices, and more particularly to an improved apparatus and method for the positioning of transducing apparatus relative to a rotating storage medium.

In the past, there have been a wide variety of approaches taken to the positioning of transducing apparatus relative to a rotating disk storage medium. These approaches involve not only differing types of positioning schemes and associated apparatus, but also differing approaches to the control thereof. Positioning schemes have included linear-linear, rotary-rotary and rotary-linear techniques. Positioning apparatus typically employed have included stepper motors, voice coil actuators and non-commutated motors. Coupling between positioning apparatus and associated transducing apparatus has been both direct and indirect. While approaches to the control of positioning apparatus were originally analog in nature, more recent designs have been of a hybrid nature, employing both analog and digital techniques.

In the linear-linear positioning scheme, both the motion produced by a motor as well as the movement experienced by a transducing assembly relative to a rotating storage medium is rectilinear in nature. The motor and the transducing assembly may be coupled in either a direct or indirect manner. In the direct coupling arrangement, the motor is directly coupled to the transducing assembly in such a manner that there exist a one-for-one relationship between motion of the motor and motion of the transducing assembly. Such an arrangement has the advantage of producing an assembly which is quite rigid, having a high resonant frequency. However, the mechanical advantage of motion produced by the motor relative to that experienced by the transducing assembly is fixed, and consequently may not be optimized. As there is generally an optimal mechanical advantage in the design of such mechanical systems, a system of levers may be employed to maximize efficacy of the system. However, use of levers necessarily results in an indirectly coupled system. While an indirectly coupled system may provide for numerous advantages, including a method to maximize efficacy of the mechanical system as well as providing for a method to simplify position decoding through the use of encoding apparatus coupled to corresponding levers, such a system introduces a flexible member in the control train between the motor and the transducer assembly. Such flexible members, however, are undesirable due to the potential introduction of error in determining position of the transducer assembly.

In a rotary-rotary positioning scheme, both the motion produced by a motor as well as the movement experienced by a transducing assembly relative to a rotating storage medium is rotary in nature. The motor and the transducing assembly may be coupled in either a direct or indirect manner. In the direct coupling arrangement, the motor is directly coupled to the transducing assembly in such a manner that there exist a one-for-one relationship between motion of the motor and motion of the transducing assembly. Such an arrangement has the advantage of producing an assembly which is quite rigid having a high resonant frequency. However, the disadvantage of such a design is that the mechanical advantage of motion produced by the motor relative to that experienced by the transducing assembly is fixed, and as a result may not be optimized. In an alternate approach, a member is typically pivoted about a selected point with the transducing apparatus on one end, and the motor coupled to the other end. The distance between the pivot point and the two alternate ends of the member, i.e., the distance between the pivot point and the transducer assembly, and the distance between the pivot point and the point of coupling to the motor, may be selected in such a way to maximize the efficacy of the mechanical system.

In a rotary-linear positioning scheme, the motion produced by a motor is rotary in nature, and the movement experienced by a transducing assembly relative to a rotating storage medium is linear in nature. Due to the difference in the type of motion produced by the motor and the desired motion of the transducing assembly, the coupling between the two is necessarily indirect. Typical of such rotary-linear positioning schemes is the familiar band-drive system frequently employed in floppy disk based systems wherein a transducer assembly is moved in a radial fashion across a surface of a rotating disk by a flexible metal band which is coupled about a pulley affixed to a shaft of a motor. Rotation of the shaft of the motor consequently results in radial movement of the transducing assembly across the surface of the disk. By changing the diameter of the pulley, the efficacy of the mechanical system may be optimized.

Positioning apparatus employed with the foregoing positioning schemes have included stepper motors, voice coils and non-commutated motors. While stepper motors have found wide application as a positioning apparatus in various positioning schemes, positioning accuracy obtainable with a stepper motor is necessarily limited due to mechanical tolerances associated with the operation of stepper motors. Consequently, tracks on the storage media must be spaced far enough apart to allow for mechanical tolerances associated with the operation of the stepper motor actuator. While these tolerances may be acceptable in relatively inexpensive floppy disk based storage systems, such limitations are highly undesirable in hard disk based system where the primary emphasis is on storage capacity. Consequently the use of stepper motors is generally avoided as a positioning apparatus in a high capacity storage system due to the resulting limitations imposed on the storage capacity of the system.

Voice coils may alternately be employed to produce either rectilinear or rotational motion. However, notwithstanding the particular design thereof, the total displacement of the moving member of the voice coils used with positioning apparatus in disk systems in the past has been limited to the difference in length between that of the magnet and the length of the coil. Consequently, while voice coils do offer an advantage in speed in operation, the limited range of motion available therefrom is a limiting consideration.

A yet further alternate approach to the production of rotational motion in rotary-rotary and rotary-linear positioning schemes is found in the use of a D.C. motor. However, due to problems associated with the control of the commutation of such motors, commutation is generally not performed. Rather, the D.C. motor is typically operated over only a limited range of rotation; frequently less than 100 degrees. A number of disadvantages follow from such an approach. By limiting the amount of useable rotational displacement of the motor, larger diameter pulleys must be attached to the rotor of the motor to achieve increased amounts of linear displacements of the transducer apparatus assembly. This necessarily requires the use of larger amounts of space, which is often at a premium in high capacity storage systems. In further addition, by operating the motor over only a limited range of angular displacement, it is more difficult to achieve a match between the rotational inertia of the motor with the reflected inertia of the head assembly. As a result thereof, it is not possible to achieve the higher efficiencies with lower power dissipation which result from achieving an optimal match between the mass of the transducer apparatus assembly and that of the D.C. motor.

In addition to the foregoing, differing approaches have been taken to the control of the transducer positioning apparatus. In particular, analog techniques were first employed. However, numerous undesirable effects including variations in component tolerances resulting from environmental effects have resulted in a shift toward control circuity employing a hybrid of analog and digital techniques. Numerous hybrid approaches have been employed in positioning the transducer assembly, with varying degrees of digital control. Typical of such approaches is a hybrid approach described in U.S. Pat. No. 4,396,959 to Harrison et al. wherein transducer apparatus position information is processed using analog apparatus including analog peak detectors and sample and hold circuits, and wherein digital position control information is converted to an analog format and subsequently processed with analog summing circuits and loop compensation networks. Further hybrid signal processing techniques may be found in U.S. Pat. No. 4,419,701 to Harrison et al. wherein analog position information is processed with an analog peak detector circuit, and wherein position control information is converted from a digital format to an analog format for processing by analog summing circuits and loop compensation networks. While such approaches do offer varying degrees of improvements in the process of transducer apparatus position determination and control, the disadvantages present with the use of analog techniques are still present to varying degrees.

Apart from hybrid hardware designs employing both analog and digital techniques, numerous control techniques have been employed with respect to controlling the position of the transducer apparatus. Such techniques have exhibited varying degrees of accuracy in controlling the position of transducer apparatus. In one control algorithm more fully described in U.S. Pat. No. 4,419,701 to Harrison el al., an open loop mode of operation is employed until the transducer apparatus is positioned at a desired destination track on the storage media, thereafter changing to a closed loop mode of operation. In other approaches where the positioning apparatus may operate in a closed loop mode, position information is determined at somewhat irregular intervals until such time as the transducer apparatus is positioned at a desired destination track. Consequently, varying degrees of reliability have been achieved in the past in the control of the positioning of transducer apparatus relative to a rotating storage medium.

From the foregoing it is clear that notwithstanding the use of differing positioning schemes, i.e., linear-linear, rotary-rotary and rotary-linear, achievement of optimal matching between the prime mover, i.e., the device producing motion, and a payload, e.g. a transducer assembly, may result in the use of some system of levers, thereby precluding advantages available from a direct coupling arrangement. This disadvantage in part results from the somewhat limited range of motion available from prime movers. Consequently, there is a need to provide for an increased range of motion by prime movers. In addition to the foregoing, there is a further need not only for more reliable approaches to the control of transducer apparatus, but also for more fully exploiting the advantages available from digital techniques in the implementation of such approaches.

In accordance with the present invention, a method and apparatus are disclosed which provides for the positioning of transducing apparatus relative to a rotating disk storage medium by use of an externally commutated D.C. motor, with the control thereof being accomplished through the use of digital techniques. Broadly stated, the control techniques employed provide not only for the commutation of the D.C. motor, but further provide for complete control over the operation of the motor in the positioning of the transducing apparatus through the use of a plurality of operational modes, the selection therebetween being determined by the distance between the present position of the transducer apparatus and the desired position therefor. The operational modes include a Calibration Mode, a High Speed Mode, a Medium Speed Mode, a Transition Mode and a Detent Mode.

Operational parameters associated with the system are determined in the calibration mode. The parameters so determined are used to adjust commands issued to the system in subsequent modes to reflect the values of the parameters measured. Consequently, a significant degree of protection against both short and long term variations in system parameters is provided. The calibration mode provides for the determination of a number of parameters. In particular, the location of commutation positions of the motor are determined. This information is subsequently used to define points at which motor commutation will occur. In addition, as the operation of the motor is controlled through digital words coupled to motor windings through digital-to-analog converters and corresponding motor drive amplifiers, offset currents associated therewith are determined. Subsequent commands to the motor are then adjusted in such a manner as to compensate for the offset currents. In further addition, as position information of the transducer apparatus assembly is detected through the use of an optical sensor, the parameters associated therewith are measured and stored for later use. In particular, the transducer apparatus assembly is moved a selected amount, and series of corresponding maximum and minimum values of the signals from optical position transducers coupled thereto are measured to determine corresponding maximum and minimum values. Thereafter, a corresponding operational zero crossing point is determined from the determined maximum and minimum values, and used in subsequent position determining operations. In further addition, the location of a preselected track on the storage medium is determined. Subsequent positioning of the transducing apparatus assembly is thereafter done with respect to the location determined for this preselected track.

Broadly stated, subsequent to the receipt of a command to move the transducing apparatus assembly from a present position to a selected position, the difference between the current position of the transducing apparatus assembly and the selected position is determined. If the difference is greater than a first preselected distance, the High Speed Mode of operation is entered, by first entereing a Medium Speed Mode, and thereafter entering the High speed Mode, as more fully discussed hereinafter. In the High Speed Mode, position of the transducing apparatus is determined with respect to the location of every other track on the rotating disk medium. Relative velocity of the transducing apparatus assembly is thereafter determined from the measured positions and the time between the measurements. Commands to the motor are thereafter produced in response to both the determined velocity and the remaining distance between the current location of the transducer apparatus assembly and the desired location. In the High Speed Mode of operation, the transducer assembly is moved at a constant maximum velocity when the distance between the current position and the desired position of the transducer assembly is greater than a first preselected distance, and according to a table of decreasing velocities thereafter.

As the transducer apparatus approaches the desired position, and the corresponding velocity has decreased to a preselected velocity, a Medium Speed Mode of operation is employed wherein position of the transducing apparatus with respect to each track location is determined. As was the case in the High Speed Mode, the velocity of the transducer assembly is adjusted according to a table of velocities as a function of distance in such a fashion to continue to decrease the velocity of the transducer apparatus. In the Medium Speed Mode, the spatial frequency of the position determinations is increased as position of the transducing apparatus is determined with respect to the location of every track on the rotating disk medium. Relative velocity of the transducing apparatus assembly is thereafter determined from the measured positions and the time between the measurements. Commands to the motor are thereafter produced in response to both the determined velocity and the remaining distance between the current location of the transducer apparatus assembly and the desired location.

As the transducer apparatus continues to approach the desired position, and the distance between the current position of the transducer apparatus assembly and the desired position becomes less than a second preselected distance, a Transition Mode of operation is entered wherein position information is determined at a higher degree of accuracy than was the case in the High or Medium Speed Modes.

When the distance between the current position of the transducer apparatus assembly and the desired position is less than a third preselected distance, a Detent Mode is entered wherein a Position Integral Differentiation control approach is employed to position the transducing apparatus assembly to the desired location.

In this regard it should be noted that in accordance with the present invention, the accuracy with which a position of the transducing apparatus relative to the rotating disk medium is determined is increased as the distance between the current position of the transducing apparatus and the desired position decreases. In particular, in the High Speed Mode, position information is determined with respect to every other track the transducing apparatus crosses in moving toward the desired destination. Thereafter, in the Medium Speed Mode, position information is determined with respect to every track the transducing apparatus crosses. Thereafter, in the Transition Mode, position information is determined with greater accuracy than was the case in the High or Medium Speed Modes, i.e., with respect to individual positions across a track. Thereafter, in the Detent Mode, position information is monitored with yet further accuracy through the additional use of not only individual positions across a track, but further with respect to position information recorded on the disk.

In addition to the foregoing, commutation of the prime mover, i.e., the apparatus moving the transducing apparatus, is further performed at appropriate times as the transducing apparatus is advanced toward the desired destination position. Commutation of the prime mover is, however, suppressed at the destination position.

In addition to the foregoing, pre-recorded position information for each track on the storage medium is further employed in the Detent Mode, as previously referenced. In particular, each track on the storage medium has associated therewith pre-recorded position information from which the position of the transducing apparatus assembly with respect to a selected track may be determined. This pre-recorded track information is sampled on each revolution of the storage medium in the Dent Mode, and position information determined therefrom is employed in positioning of the transducing apparatus assembly with respect to the desired track.

FIG. 1 is a broad functional block diagram of apparatus in accordance with the present invention.

FIG. 2 is a schematic diagram of an implementation of a Gated Integrator employed in the present invention.

FIG. 3 illustrates positional information of servo bursts used in the present invention.

FIGS. 4A and 4B are broad functional illustrations of externally commutated motors capable of producing rotational and linear displacement of transducing apparatus in accordance with the present invention.

FIGS. 5A thru 5E illustrates functional operations associated with the High Speed Mode, the Medium Mode, the Transition Mode and Detent Mode of operation.

In accordance with the present invention, a method and apparatus are disclosed which provide for an improved transducer position control system for disk storage equipment employing an externally commutated motor and digital control techniques.

FIG. 1 is a functional block diagram of apparatus in accordance with the present invention. Referring to FIG. 1, Microprocessor 10 functions as a central control element to orchestrate the operations of the transducer positioning apparatus, as will be more fully described hereinafter. Microprocessor 10 communicates with devices external thereto by information passing through a number of functional terminals associated therewith. Microprocessor 10 selects Digital-to-Analog Converter 40 for receipt of digital information placed on data bus terminal 14 by placing an appropriate signal on terminal 12. In a similar fashion, Microprocessor 10 selects Digital-to-Analog Converter 50 for receipt of digital information placed on data bus terminal 14 by placing an appropriate signal on terminal 16. Microprocessor 10 controls the operation of Analog Multiplexer 90 by control signals placed on terminal 18. Microprocessor 10 controls the operation of Analog-to-Digital Converter 110 by control signals placed on terminal 20. Microprocessor 10 receives information from Analog-to-Digital Converter 110 on terminal 22. Microprocessor 10 places control information for Gated Integrator 130 on terminal 24. Microprocessor 10 receives periodic angular position information defining a selected angular position of a disk 122 on terminal 26. Microprocessor 10 receives information from and communicates to other external devices by information coupled to terminal 28. It is understood that the foregoing discussion of particular terminals through which Microprocessor 10 communicates with other devices was in terms of functionality of the operations being performed; the exact nature and manner in which a particular type of microprocessor device would in fact accomplish the aforedescribed communication would be determined by the particular type of device selected, the details as to device interfacing being apparent to one of ordinary skill in the art. In the preferred embodiment, Microprocessor 10 was implemented by a model 6301 microprocessor manufactured by Hitachi Corporation of Tokyo, Japan. A timer 30 such as a conventional programmable timer is connected to microprocessor 10 to provide the positioning system with a timing capability. Alternatively, the timer could be an integral part of microprocessor 10.

Digital-to-Analog Converter 40 has associated therewith a select terminal 42, an input terminal 44 and an output terminal 46. Digital-to-Analog Converter 40, responsive to a select signal present on select terminal 42, and a digital word on terminal 44, functions to couple an analog value to terminal 46 having a magnitude proportional to the digital word coupled to terminal 44.

Digital-to-Analog Converter 50 has associated therewith a select terminal 52, an input terminal 54 and an output terminal 56, and functions in an identical manner as Digital-to-Analog Converter 40. While any of a wide variety of digital-to-analog converter devices may be used, a model AD7528 Dual Analog-to-Digital Converter manufactured by Analog Devices of Norwood, Mass. was used in the preferred embodiment to implement the functions of Digital-to-Analog Converters 40 and 50.

Motor 80 is a two pole, 2 phase D.C. motor having motor windings 82 and 84 associated therewith. In this regard it is to be understood that while a two pole, 2 phase externally commutated brushless D.C. motor was used in the preferred embodiment, and is illustrated in FIG. 1, other types of externally commutated positioning apparatus may likewise be employed, including externally commutated voice coil motors. In a similar fashion, while the externally commutated positioning apparatus was of a two pole design in the preferred embodiment, any number of poles could likewise be employed in accordance with the teachings of the present invention. Such other types of externally commutated positioning apparatus having a plurality of poles would be apparent to one of ordinary skill in the art and are to be considered to be within the teachings of the present invention.

Motor Drive Amplifier 60 has associated therewith input terminal 62 and output terminals 64, and functions responsive to an analog signal coupled to input terminal 62 to produce a corresponding analog output signal on output terminals 64 suitable for driving a winding of motor 80.

Motor Drive Amplifier 70 has associated therewith input terminal 72 and output terminals 74, and functions in an identical manner as Motor Drive Amplifier 60. While the functions of Motor Drive Amplifiers 60 and 70 may be implemented by any of a wide variety of devices, model L272 Dual Power Operational Amplifiers manufactured by SGS Semiconductor Corporation of Phoenix, Ariz. were employed in the preferred embodiment.

Analog Multiplexer 90 has associated therewith select terminal 92, analog input terminals 94, 96 and 98, and analog output terminal 100. Analog Multiplexer 90 responsive to a select signal coupled to select terminal 92 operates to couple an analog signal present on one of analog input terminals 94, 96 or 98 to analog output terminal 100. The particular analog input terminal which is coupled to analog output terminal 100 is determined by the signal coupled to select terminal 92. While the functions of Analog Multiplexer 90 may be implemented by any of a wide variety of analog multiplexer devices, a model 4051 Analog Multiplexer manufactured by National Semiconductor of Santa Clara, Calif. was used in the preferred embodiment.

Analog-to-Digital Converter 110 has associated therewith select terminal 112, analog input terminal 114 and digital output terminal 116, and functions responsive to a select signal coupled to select terminal 112 to convert the magnitude of the analog signal which is present on analog input terminal 114 to a corresponding digital value on digital output terminal 116. While the functions of Analog-to-Digital Converter 110 may be implemented by any of a wide variety of analog-to-digital devices, a model ADC0820 Analog-to-Digital Converter manufactured by National Semiconductor of Santa Clara, Calif. was used in the preferred embodiment of the present invention.

Disk 122 serves as a storage medium for information, and is rotated by Spindle motor 124. Transducer apparatus 120 functions to transfer information onto as well as retrieve information from the surface of disk 122. Transducer apparatus 120 is mechanically coupled to the rotor of Motor 80 in such a fashion that Transducer apparatus 120 may be moved relative to selected positions on the surface of disk 122 responsive to the rotation of the rotor of Motor 80. In this regard, it is understood that the movement of Transducer assembly 120 across the surface of Disk 122 may be according to either the linear-linear, rotary-rotary or rotary-linear positioning schemes, using either a direct or indirect coupling arrangements. In the preferred embodiment, Transducer assembly 120 was indirectly coupled to the rotor of Motor 80 by a standard band and pulley coupling apparatus, well known to one of ordinary skill in the art.

Preamplifier 131 has associated therewith terminals 133, 135 and 137, and operates in conjunction with Transducer apparatus 120 in the electrical coupling of information between Transducer apparatus 120, External Information Processing Apparatus 139 and Gated Integrator 130. In particular, Preamplifier 131 functions to transfer information between Transducer apparatus 120 and External Information Processing Apparatus 131, and from Transducer assembly 120 to Gated Integrator 130, as more fully described hereinafter. While any of a wide variety of preamplifier devices well known to one of ordinary skill in the art may be used, a model 501 Preamplifier manufactured by Silicon Systems, Inc of Tustin, CA., was used in the preferred embodiment.

Disk 122 has associated therewith Index Sensor 126 which functions to produce a signal once per revolution of Disk 122 responsive to the passage of a selected angular position thereby.

Gated Integrator 130 has associated therewith input terminal 132, control terminal 134 and output terminal 136, and operates responsive to a signal coupled to control terminal 134 to rectify and thereafter electrically integrate a signal coupled to input terminal 132, producing the corresponding output signal on output terminal 136, as will be more fully discussed hereinafter.

Encoder 140 is mechanically coupled to Motor 80, and consequently functions to provide information with respect to the position of transducer assembly 120 with respect to the surface of disk 122. In particular, responsive to the rotation of the shaft of Motor 80, Encoder 140 produces two periodic signals, one on terminal 142 and the other on terminal 144. The periodic signals are typically sine and cosine, or triangular waveforms displaced with respect to each other by ninety degrees. The signals produced by Encoder 140 are used to determine the relative position of Transducer apparatus 120 with respect to disk 122. In the preferred embodiment, each zero crossing of a selected signal from Encoder 140 defines the approximate location of a track on disk 122. As each of the signals produced by Encoder 140 typically has a relative linear and non-linear portion, both signals are typically employed for the accurate determination of position information. In particular, when a non-linear portion of one signal is encountered, the alternate signal is typically in the relatively linear portion, and is used for the determination of position information, as more fully discussed hereinafter. Consequently, by switching between the two signals as necessary, continuously accurate position information of Transducer assembly 120 with respect to the surface of Disk 122 may be determined. Encoder 140 may be any of a wide variety of encoding devices. The number of cycles of the periodic signals produced per revolution by Encoder 140 in response to rotation of the shaft of Motor 80 140 is referred to as the pitch. Typical pitches may be 1000 cycles per 360 degree rotation of the shaft of motor 80. In the preferred embodiment, Encoder 140 has a pitch of 600, and is manufactured by Computer Optical Products of Chatsworth, Calif.

The foregoing described apparatus is configured in the following manner. Terminal 12, 14, 16, 18, 20, 22 and 24 of Microprocessor 10 is coupled to terminal 42 of Digital-to-Analog Converter 40, terminals 44 and 54 of Digital-to-Analog Converters 40 and 50 respectively, terminal 52 of Digital-to-Analog Converter 50, terminal 92 of Analog Multiplexer 90, terminal 112 of Analog-to-Digital Converter 110, terminal 116 of Analog-to-digital Converter 116, and terminal 134 of Full Wave Rectifier and Gated Integrator 130, respectively. The signal from Index Sensor 126 is coupled to terminal 26 of Microprocessor 10. Microprocessor 10 communicates with respect to external positioning commands through terminal 28. Terminal 46 of Digital-to-Analog Converters 40 is coupled to terminal 62 of Motor Drive Amplifier 60. Terminals 64 of Motor Drive Amplifier 60 are coupled to winding 82 of motor 80. In a similar fashion, terminal 56 of Digital-to-Analog Converter 50 is coupled to terminal 72 of Motor Drive Amplifier 70. Terminals 74 of Motor Drive Amplifier 70 are coupled to winding 84 of Motor 80. Terminals 142 and 144 of Encoder 140 are coupled to terminals 94 and 96, respectively, of Analog Multiplexer 90. Transducer apparatus 120 is coupled to terminal 133 of Preamplifier 131. Preamplifier 131 communicates with External Information Processing Apparatus 139 through terminal 135. Terminal 137 of Preamplifier 131 is coupled to terminal 132 of Full Wave Rectifier & Gated Integrator 130. Terminal 136 of Full Wave Rectifier and Gated Integrator 130 is coupled to terminal 98 of Analog Multiplexer 90.

FIG. 2 illustrates the apparatus employed in the preferred embodiment to implement Gated Integrator 130. Referring to FIG. 2, resistor 150 couples the signal from terminal 24 of Microprocessor 10 to the base of transistor 152, and corresponds to terminal 134 of Gated Integrator 130 (FIG. 1). Resistor 154 couples the base of transistor 152 to the emitter thereof. Resistor 156 and capacitor 158 are coupled between the emitter and collector of transistor 152. Resistor 160 couples the collector of transistor 152 to the base of transistor 162. The emitter of transistor 152 and the collector of transistor 162 are coupled to a source of positive potential. The emitter of transistor 162 is coupled to a ground reference by resistor 164. The respective collectors of transistors 166 and 168 are both coupled to the collector of transistor 152, and the emitters of both transistors 166 and 168 are likewise coupled together. Resistor 170 couples the emitters of transistors 166 and 168 to a ground reference. The base of transistor 166 is coupled to the collector of transistor 176 through resistor 174, and the base of transistor 168 is coupled to the collector of transistor 176 through resistor 172. The collector of transistor 176 is coupled to a source of positive potential through resistor 178. The base of transistor 176 is coupled to the collector thereof. The emitter of transistor 176 is coupled to ground reference. The signal from transducer assembly 120 is coupled to the base of transistors 166 and 168 by coupling capacitors 180 and 182, and together correspond to terminal 132 of Gated Integrator 130 (FIG. 1). The emitter of transistor 162 is coupled to terminal 98 of Analog Multiplexer 90, and corresponds to terminal 136 of Gated Integrator 120 (FIG. 1).

The apparatus of FIG. 2 operates in the following manner. Responsive to a signal from terminal 24 of Microprocessor 10, transistor 152 operates to discharge capacitor 158. Thereafter, a signal from transducer assembly 120 operates to control the charging of capacitor 158 by the operation of transistors 166 and 168. The potential to which capacitor 158 charges is coupled to terminal 98 of Analog Multiplexer 90 by the emitter follower combination of transistor 162 and resistor 164. Consequently, subsequent to a signal from terminal 24 of Microprocessor 10, Gated Integrator 130 operates to electrically integrate the signal from transducer assembly 120, and couple the value thereof to Analog Multiplexer 90, where it is subsequently coupled to Microprocessor 10 through Analog-to-digital Converter 110, as more fully discussed hereinafter.

Information is typically stored across the surface of disk 122 in concentric bands referred to as tracks. Information used to define a precise track location is placed on each track, as illustrated in FIG. 3 with respect to a single track 200. Referring now to FIG. 3, a segment of a track 200 on the surface of Disk 122 is illustrated. For the purposes of the present discussion, it is assumed that Transducing apparatus 120 is positioned directly over track 200. Two discrete packets of information, referred to as servo bursts, are recorded on each track on the surface of disk 122, indicated as servo burst 202 and servo burst 204 in FIG. 3. Each of the discrete servo bursts 202 and 204 are positioned adjacent to and on opposite sides of the center of tract 200, and displaced from each other by a selected distance 206. As a result of the foregoing, each of servo bursts 202 and 204 will pass under transducer apparatus 120 at different points in time. The maximum amplitude of the resulting signal in transducer 120 in response to the passage of each of servo bursts 202 and 204 is measured by the operation of Gated Integrator 130, compared by Microprocessor 10, and used to correct the positioning of Transducer apparatus 120 with respect to track 200 in the following manner. If Transducer apparatus 120 is properly centered on track 200, the maximum amplitudes produced by servo burst 202 and 204 passing thereunder will be equal. However, if Transducing apparatus 120 is not correctly centered over track 200, there will be a difference in the maximum amplitudes produced by each of servo burst 204 and 206 when read by Transducer apparatus 120. The detection of a difference in amplitudes is thereafter used by Microprocessor 10 to properly position transducer assembly 120 with respect to track 200, as will be more fully discussed hereinafter.

The foregoing operates in the following manner. Referring once again to FIG. 1, Index Sensor 122 operates to produce a signal indicating the passage of a selected angular position on disk 122 thereby. This identified position serves as a reference position for information stored on individual tracks on disk 122. In particular, servo bursts 202 and 204 are recorded on disk 122 in such a fashion to pass under transducer 120 immediately following the passage of the reference position identified by the signal produced by Index Sensor 122. In operation, immediately subsequent to the receipt of the signal from Index Sensor 122 by Microprocessor 10 on terminal 26, Microprocessor 10 effects the resetting of capacitor 158 of Gated Integrator 130 (FIG. 2) by the generation of a signal to terminal 24 and coupled to terminal 136 of Gated Integrator. Thereafter, the signals produced in Transducer 120 (FIG. 3) by the passage of servo burst 202 thereunder and the subsequent integration thereof is coupled from Gated Integrator terminal 136 to Analog Multiplexer terminal 98. Microprocessor 10, by placing an appropriate signal on terminal 18 thereof effects the selection of the signal from Gated Integrator 130 for coupling to Analog-to-Digital Converter 110. Thereafter, in response to an appropriate signal from terminal 20 of Microprocessor 10 coupled to terminal 112 of Analog-to-Digital Converter 110, the magnitude of the amplitude of the signal produced by servo burst 202 will be converted to a corresponding digital value, and coupled to terminal 22 of Microprocessor 10. The foregoing process will then be repeated with respect to servo burst 204. Microprocessor 10 will thereafter compare the amplitudes resulting from the passage of servo bursts 202 and 204: equality of the two digital words indicating the correct positioning of transducer 120 (FIG. 2) over track 200, and inequality indicating incorrect positioning. Consequently, the position of transducer assembly 120 with respect to a track on Disk 122 may be precisely determined, and the position thereof subsequently corrected as necessary, as more fully described hereinafter. It will be understood in this regard that while the previously discussed zero crossing on a selected one of the signals produced by Encoder 140 is intended to identify the approximate location of tracks, the effect of a number of practical considerations results in a variation in the actual location of the tracks on the surface of Disk 122. In this regard it is recognized by those skilled in the art that thermal effects will result in the change of location of tracks on the surface of Disk 122. Consequently, the position information determined from the aforedescribed servo bursts is used to further qualify position information determined from the signals produced by Encoder 140, as will be more fully described hereinafter.

The position motor used in the preferred embodiment of the present invention is a two pole, two phase torquer motor, as functionally illustrated in FIG. 4A. Referring to FIG. 4A, a bar magnet 210 is positioned in the center of a circularly shaped magnetic permeable material, and pivoted for rotation at its center point 212. Terminals A 214 and B 216 represent terminals for windings about the circularly shaped magnetic permeable material. In particular, a first winding is wound around a first half, and a second winding about a second half of the circularly shaped magnetic permeable material, with the respective ends thereof connected to terminals A 214 and B 216. The windings are wound in such a manner that when a current is passed through the windings from terminal A 214 to terminal B 216, magnet 210 will rotate in a first direction approximately 90 degrees from an initial position approximately midway between the end points, i.e., terminal A 214 and B 216 of the windings. If the current is then reversed, magnet 210 will then rotate approximately 90 degrees in the opposite direction. In a similar fashion, terminals C 218 and D 220 likewise represent terminals for similar winding about the circular shaped magnetic permeable material, and are displaced from the previously discussed windings by 90 degrees. The windings connected to terminals C 218 and D 220 function in a similar manner: current in a first direction from terminal C 218 to terminal D 220 will result in rotation of magnet 210 approximately 90 degrees in a first direction from an initial position approximately midway between the end points, i.e., terminal C 218 and D 220, of the windings. If the current is then reversed, magnet 210 will then rotate approximately 90 degrees in the opposite direction. The first set of windings about the circularly shapped magnetic permeable material which connect to terminals A 214 and B 216 correspond to a single coil 82 or 84 of FIG. 1, and the second set of windings which connect to terminals C 218 and D 220 correspond to the other single coil, 84 or 82 of FIG. 1.

While the foregoing has broadly described the particular type of motor used in the preferred embodiment, it is to be understood that the teachings of the present invention may likewise be applied to other types of commutated motors, including a linear voice coil motor. FIG. 4B broadly illustrates a commutated linear voice coil motor which could be used in accordance with the teachings of the present invention. Referring to FIG. 4B, two magnets 222 and 224 are positioned adjacent to pole pieces 226 and 228, respectively. In this regard, like poles are positioned against the respective pole pieces, i.e., a north pole of magnet 222 is positioned adjacent to pole piece 226, and a north pole of magnet 224 is positioned adjacent to pole piece 228. The opposite poles of magnets 222 and 224 are each positioned adjacent to a common member 230, which further has placed adjacent to it a member 232. The members 226, 228, 230 and 232 operate to complete the respective magnetic circuits for magnets 222 and 224: the magnetic circuit associated with magnet 222 being comprised of members 226, 232 and 230, and in a similar manner, the magnetic circuit associated with magnet 224 being comprised of members 228, 232 and 230. A further member 234 having windings therearound is arranged for motion over member 232 under the influence of the magnetic fields produced by magnets 222 and 224 and a current flowing in the windings associated therewith, in a manner well known to those skilled in the art. Member 234 is further coupled to a payload 236. In accordance with the teachings of the present invention, multiple windings may be placed around member 234. In particular, a first winding terminating at points A 238 and B 240, is positioned at a selected position along the length of member 234; e.g., adjacent to one end of member 234 and extending approximately 3/4 of the length thereof. A second winding terminating at points C 242 and D 244 is positioned at a selected position along the length of member 234; e.g., adjacent to the opposite end of member 234, and extending approximately 3/4 of the length thereof. In a similar manner as previously discussed with respect to the motor of FIG. 4A, the windings associated with member 234 may likewise be commutated in accordance with the teachings of the present invention. Consequently, the teachings of the present invention may be applied by one of ordinary skill in the art to a broad range of commutated motors, and are not to be considered as limited to the particular type of motor described with respect to the preferred embodiment herein.

The characteristics of the two pole, two phase torquer motor used in the preferred embodiment previously discussed are those of a four step per revolution stepper motor, as will be understood by those skilled in the art. When a single phase alone of the two phase winding of the motor is excited with a fixed level of current, motor torque is developed, which tends to cause rotation of the rotor toward a magnetic null or zero torque position. Reversal of the polarity in the winding will cause the motor to seek a magnetic null 180 degrees away from the first null. Excitation of the second winding alone will similarly cause two magnetic nulls to be found, again at 180 degrees from each other, and mutually 90 degrees away from the magnetic nulls defined by the first winding of the motor. It is thus seen that the two windings define four magnetic null positions for the motor.

With a single winding excited with a fixed level of current, the motor will exhibit approximately a constant torque tending to rotate the rotor toward the magnetic null. This constant torque will be exhibited over a significant percentage of the 180 degrees between the null and the anti-null position.

Since the two pole two phase motor, having four magnetic null positions, acts as a four step per revolution stepper motor, the motor can be caused to rotate in 90 degree steps in a known direction by alternatively exciting each of the two windings with successively opposite polarities of current. As each winding is excited with a fixed level of current and polarity, the motor will seek to one of its four magnetic positions. Excitation of the other winding will then cause the motor to perform another 90 degree step in a known direction. Progressive sequential excitation of the two phases and polarities will thus cause the motor to move in 90 degree steps in a known direction, independent of the initial position of the motor.

Broadly stated, the apparatus of FIG. 1 operates to position transducer 120 to a selected position on disk 122 in the following manner. Referring once again to FIG. 1, the apparatus first performs a calibration sequence to determine current values of parameters associated with the system. In particular, the commutation points for Motor 80 are determined, the maximum and minimum values for both periodic signals produced by Encoder 140 are determined, the offset currents associated with the operation of the combinations of Digital-to-Analog Converter 40 and Motor Drive Amplifier 60, and Digital-to-Analog Converter 52 and Motor Drive Amplifier 70 are determined, and the location of a selected track on the surface of disk 122 is determined.

Initially, Motor 80, Encoder 140 and Transducer apparatus 120 are in an unknown position. It is therefore necessary to cause the motor to move to a known position from which successive programmed motions may be initiated. Mechanical stops placed at both the outer and inner periphery of Disk 122 operate to limit motion of Transducer assembly 120 to selected positions at the outer and inner periphery of Disk 122. Motor 80, under control of Microprocessor 10, operates through successive excitation of windings 82 and 84 of Motor 80 through the associated Motor Drive Amplifiers 60 and 70, and Digital-to-Analog Converters 40 and 50 to position Transducer assembly 120 against the outer mechanical stop. However, because of the possibility that the unknown initial position of Motor 80 may in fact already be against the inner or outer stop, it is necessary to confirm first that motion has in fact been initiated, and subsequently, that motion has ceased to occur when the motor contacts the outer stop. This is accomplished by Microprocessor 10 monitoring the outputs 142 and 144 of Encoder 140 through Analog Multiplexer 90 and Analog-to-Digital Converter 110 on terminal 22 of Microprocessor 10.

After Transducer assembly 120 has been positioned against the outer mechanical stop as above described through the observation of the signals from Encoder 140, Motor 80 will be in a known rotational position. From this known rotational position, programmed sequences of motion can be initiated to achieve desired new positions of Motor 80, Encoder 140 and Transducer 120, as more fully described hereinafter.

Calibration data on the output signals from terminals 142 and 144 of Encoder 140 is next determined. By successive excitation of the two phases of Motor 80 in a stepping mode, Transducer apparatus 120 is caused to move from the known position at the outer mechanical stop to the inner mechanical stop. While this motion is occurring, Microprocessor 10, operating through Analog Multiplexer 90 and Analog-to-Digital Converter 110 samples the outputs from terminals 142 and 144 of Encoder 140, and records the minimum and maximum values thereof. In the preferred embodiment, continuous motion of Transducer apparatus 120 across Disk 122 will produce a sine signal on terminal 142 and a cosine signal on terminal 144 of Encoder 140. The stored maximum and minimum digital words consequently represent the maximum and minimum values of the corresponding sine and cosine signal outputs from Encoder 140. The difference between the maximum and minumum values for both the sine and cosine signals respectively, define peak-to-peak output values from Encoder 140 for the two respective signals, i.e. sine and cosine, from terminals 142 and 144, and therefore a scale factor which allows Microprocessor 10 to determine an exact angular position of Encoder 140 at any future time.

An average value for each of the sine and cosine signals, obtained by adding the respective maximum and minimum digital words for each of the sine and cosine signals, and dividing by two in Microprocessor 10, defines a mid-point on each of the sine and cosine signals, which is a zero cross-over point for the two respective signals. Each zero cross over point of the sine signal defines a nominal location of a center of a magnetic track on Disk 122.

Subsequent to completing the encoder calibration procedure as above described, Motor 80 is caused to return Transducing apparatus 120 again to the outer mechanical stop, in the same manner as above described. Thereafter zero offset errors associated with the operation of Digital-to-Analog Converters 40 and 50 and Motor Drive Amplifiers 60 and 70 are determined. In particular, the exact digital word required to be coupled to each of Digital-to-Analog Converters 40 and 50 to achieve zero current out of the associated Motor Drive Amplifier is determined. The foregoing is accomplished by sequentially exciting each of windings 82 and 84 of Motor 80 to effect the positioning of Transducer assembly 120 in the approximate center of the area on Disk 122 wherein data is stored. One of the windings of Motor 80 is then energized and Motor 80 is allowed to seek the associated magnetic null. The opposite winding is then energized and a position servo loop described in more detail hereinafter with respect to the Detent Mode of operation, is operated to cause Transducer assembly 120 to seek one of the zero crossing positions of the sine wave signal from terminal 142 of Encoder 140. The position thus established is exactly on the nominal center of one of the tracks on Disk 122, and stable position control is established. With the motor encoder and transducer at controlled rest, there can be no torque developed by Motor 80, and therefore no current in windings 82 or 84. With this zero current condition established, the digital word at the input of Digital-to-Analog Converter 40 or 50 which produces this zero current condition is noted and stored by Microprocessor 10. The difference between this word and the nominal zero current command word are determined and stored.

During the determination of this offset correction word at the input of the respective digital-to-analog converters, Microprocessor 10 causes Transducer assembly 120 to "dither" slightly about its nominal rest position in order to minimize the effects of friction which might otherwise distort the measurement of this offset word. In the preferred embodiment a total of 400 readings of the input words to the digital-to-analog converters are taken and averaged by Microprocessor 10 to determine an average offset value.

After the offset compensation value for one digital-to-analog converter has been determined, as above described, that same winding is then fully energized, causing Motor 80 and Transducer apparatus 120 to position to the magnetic null position defined by that winding. The other winding is then energized in the position control loop as above described above, and the offset digital word required for the digital-to-analog converter controlling the second winding is similarly determined.

In order to maintain constant torque as a function of current from Motor 80, it is necessary to switch, or commutate, between the two windings 82 and 84 at a position approximately halfway between the magnetic nulls defined by the two windings, or at 45 degrees between the four magnetic nulls. The purpose of the commutation determination procedure is to identify the location of the four magnetic nulls, angularly, as measured by Encoder 140. With these measurements, the optimum commutation point for Motor 80 is then defined as being half way between the measured magnetic null angular positions.

A counter, hereinafter referred to as a Commutation Counter, is employed in Microprocessor 10 in the commutation process. The Commutation Counter increments or decrements responsive to each crossing of a track by Transducer assembly 120; the Commutation Counter increments on each zero crossing of the sine signal resulting from motion of Transducer apparatus 120 in a first direction of motion with respect to Disk 122, and decrements responsive to each crossing of a track by Transducer assembly 120 in the opposite direction of motion. In response to motion of Transducing apparatus 120 in the first direction, the Commutation Counter is arranged to count up to a preselected maximum value, and thereafter to reset to zero, and repeat the counting process once again. In response to motion of Transducing apparatus 120 in the opposite direction, the Commutation Counter is arranged to count down to zero, and thereafter reset to the preselected maximum value, and again continue the decrementing process. Commutation of the windings of Motor 80 occur each time the count in the Commutation Counter changes from the maximum value to zero, or from zero to the maximum value.

After completion of the determination of the offset values of the digital-to-analog converters and the associated motor drive amplifiers as above discussed, Motor 80 is again energized in a stepper mode, and caused once again to go to the outer mechanical stop. After Motor 80 has come to rest against the outer mechanical stop, Motor 80 is caused to step inward in a location toward the center of the area on which data is to be stored on Disk 122. One of the two windings 82 or 84 of Motor 80 is then energized and Motor 80 is allowed to rotate to the corresponding magnetic null position. As has been described above, this magnetic null position is approximately half way between the optimum commutation points between the two windings. With Motor 80 in this magnetic null position, a commutation counter in Microprocessor 10 is set to a value equal to one-half of the previous discussed preselected maximum value. In the preferred embodiment, the preselected maximum value is 300, hence the Commutation Counter is set to a value of 150. The Commutation Counter is arranged that it will overflow at a count of 300, resetting to 0, and underflow at a count of 0, resetting to 300. The input to this Commutation Counter is derived from the zero crossing signals obtained by the sine output of Encoder 140. The counter continually counts track crossings and causes commutation between the two windings to occur whenever either an underflow or overflow occurs in the Commutation Counter.

The final task to be performed in the calibration mode is to position Transducer apparatus 120 exactly over track number 0, i.e., the first track used for the storage of data. In this regard it will be understood that in the preferred embodiment, track 0 corresponds to the first track on which data information will be stored. However, in the preferred embodiment there are further tracks positioned between track 0 and the outer mechanical stop which are not used to for the storage of data information, and will be referred hereinafter as tracks having negative track numbers, e.g., tracks -1, -2, etc. with track -1 being adjacent to track 0. As noted above, at the completion of the initialization of the Commutation Counter, Transducer assembly 120 is positioned over the area on Disk 122 used for the storage of data information. Prior to the initiation of the commutation calibration routine, a counter in Microprocessor 10 has been reset to 0 with Motor 80 positioned against the outer mechanical stop. This counter will be hereinafter referred to as the Track Crossing Counter. As Transducer assembly 120 is moved inward, Track Crossing Counter is incremented as each track crossing is indicated by the output of the sine signal from terminal 142 of Encoder 140. When the Commutation Counter initialization is complete, the Track Crossing Counter contains a number which is equal to the total number of tracks crossed from the outer mechanical stop to the current location of Transducer assembly 120.

During the manufacturing process, servo burst information previously described with respect to FIG. 3 (servo bursts 202 and 204) was permanently prerecorded on every track of Disk 122, with the exception of the tracks having a negative track number. Thus all tracks beginning with track zero and inward toward the center of Disk 122 have prerecorded both servo bursts 202 and 204 (FIG. 3). All tracks having a negative track number, however, have prerecorded only one of the two servo burst, e.g., servo burst 202.

Microprocessor 10 issues a command to a position control servo loop, more fully described hereinafter, which causes Motor 80 to position Transducer assembly 120 over a selected track having a negative number inward from the outer mechanical stop. As this track will contain only one of the servo bursts, examination of servo burst signals therefrom by Microprocessor 10 through Preamplifier 131, Gated Integrator 130, Analog Multiplexer 90 and Analog-to-Digital Converter 110 (FIG. 1) will indicate the absence of one of the servo bursts, i.e., there will be a large difference between the amplitudes determined for each of the servo bursts. When this large difference is found, indicating the lack of both servo bursts, Microprocessor 10 issues a command to the servo loop which causes Transducer assembly 120 to be moved inward by one track. The servo burst comparison is again made, and if a large difference is again found, another inward step is initiated. This process continues until Microprocessor 10 finds the first track on which both servo bursts exist. This is the desired track 0. When track 0 has been identified, the value in the Track Crossing Counter is noted and stored by Microprocessor 10. At all later times, the track number can then be determined by subtracting this prerecorded value from the value in the Track Cross Counter.

If, by reason of some malfunction or other difficulty, Microprocessor 10 does not find a missing servo burst, the process is repeated a selected number of times. If Microprocessor 10 still does not find a missing servo burst, the entire calibration process is aborted, and an operator notified by a signal from terminal 28 of Microprocessor 10.

Microprocessor 10 makes use of three counters in the positioning of Transducer apparatus 120 to a desired location on Disk 122: a Track Crossing Counter, a Commutation Counter and a Tracks To Go Counter.

The Track Crossing Counter has been previously discussed with regard to the calibration operations, and is a counter which increments and decrements based on zero crossings of the sine signal from terminal 142 of Encoder 140 (FIG. 1). Every zero crossing of the sine signal from terminal 142 defines an approximate track center location. Track Crossing Counter maintains a continuous record of the location of Transducer apparatus 120 relative to the tracks of Disk 122 by incrementing and decrementing as the tracks are crossed in the calibration mode. The determination to increment or decrement is made based on the direction of motion as the zero crossings of the sine signal occur.

The Tracks To Go Counter is a counter which maintains a record of how far away Transducing assembly 120 is from a desired track. External Commands coupled to terminal 28 of Microprocessor 10 (FIG. 1) cause the Tracks To Go Counter to increment, and track crossings determined from zero crossings of the sine signal from terminal 142 of Encoder 140 are used to decrement the Tracks To Go Counter.

In the calibration procedure previously described, the Tracks Crossing Counter alone is used to maintain a continuous record of the location of Transducer apparatus 120. However, in the various seek modes to be more fully discussed hereinafter, both the Track Crossing Counter and the Tracks To Go Counter are employed. At the start of a seek operation, i.e., moving Transducing apparatus 120 from a current position to a desired position, the Track Crossing Counter is updated to the number of the destination track, while the To Go Counter maintains the actual position relative to the destination, being updated by track crossing signals, i.e., zero crossings of the sine signal from terminal 142 of Encoder 140.

The Commutation Counter is used to determine when to commutate the windings 82 and 84 associated with Motor 80, through the associated Digital-To-Analog Converters 40 and 50, and Motor Drive Amplifiers 60 and 70. The Commutation Counter increments and decrements in the same manner as the Tracks To Go Counter. The Commutation Counter, however, is preset during the calibration process and overflows or underflows at a preselected values, with commutation occurring with each over or underflow. Thus, the Commutation Counter continually maintains a record of which of the windings of Motor is desired to be energized.

A control servo loop consists of Microprocessor 10, Digital-to-Analog Converters 40 and 50, Motor Drive Amplifiers 60 and 70, the windings 82 and 84 of Motor 80, Encoder 140 and Transducing apparatus 120. The output of Encoder 140 provides the feedback signal through Analog Multiplexer 90 and Analog-to-Digital Converter 110. As has been previously discussed, the two output signals from terminals 142 and 144 of Encoder 140 provide continuous instantaneous position information of Transducer apparatus 120. The velocity with which Transducer assembly 120 is moving relative to the Disk 122 is determined by measuring the time between successive measurements of displacement signals from Encoder 140. Microprocessor 10 further implements a digital integrator, whose input is a calculated position error signal derived from the output of Encoder 140, as more fully discussed hereinafter with respect to the Detent mode of operation.

It will be understood by those skilled in the art that Microprocessor 10 operates as a multiplexed data processor wherein all necessary computations for the control of the disk drive are performed with and through Microprocessor 10, according to a permanently recorded control program and associated information tables, as more fully discussed hereinafter. As it is inherent in the operation of Microprocessor 10 that only one calculation may be performed at any single time, control calculations are accomplished by sequentially performing the many calculations required for the complete control process. The speed at which Microprocessor 10 is capable of performing the required calculations is sufficiently high that Microprocessor 10 is able to perform all of the necessary calculations required with respect to the control process by multiplexing, i.e., by successively performing the various calculations in sequence.

As previously discussed, four control modes are employed according to the present invention in the positioning of Transducer apparatus 120 from a current position to a desired position: a High Speed Mode, a Medium Speed Mode, a transition Mode and a Detent Mode. As the calculations required in each of the four control modes are different, the the time required for each also differ. Consequently, the manner in which each of the control modes operates are based upon the required calculations times, as will be more fully discussed hereinafter.

The four modes of operation provide for the orderly positioning of Transducing apparatus 120 from a current position to a desired position. Consequently, each of the four modes have unique operational characteristics selected to accomplish the desired positioning task.

Broadly stated, the High Speed Mode of operation relates to the management of the positioning of Transducing apparatus 120 when it is moving at a velocity greater than a preselected velocity. In the preferred embodiment, the first preselected distance is 108 tracks, and the preselected velocity is 6.7 inches per second. In the High Speed Mode of operation, if the distance from the desired destination is greater than the first preselected distance, the High Speed Mode of operation requires Transducing apparatus 120 to move toward the desired position at a preselected maximum velocity which, in the preferred embodiment is 10 inches per second. If the distance is less than the first preselected distance, the High Speed Mode of operation requires Transducing apparatus 120 to move toward the desired position at a velocity which is determined according to the distance remaining to travel. This distance-velocity relationship is according to the well known fractional power relationship, the precise fractional power being empirically determined from the mechanical and electrical characteristics of a particular design, as is well known to one of ordinary skill in the art. In the High Speed Mode of operation, position of Transducing apparatus 120 is monitored with respect to every other track crossing on Disk 122, i.e., with respect to every other zero crossing of the sine signal.

Broadly stated, the Medium Speed Mode of operation relates to the management of the positioning of Transducing apparatus 120 when it is less than the first preselected distance from the desired destination (108 tracks in the preferred embodiment), but greater than a second preselected distance (3 tracks in the preferred embodiment), and when the velocity of Transducing apparatus is less than the preselected velocity (6.7 inches per second in the preferred embodiment). In the Medium Speed Mode of operation, the desired velocity of Transducing apparatus 120 is again determined by the remaining distance from the desired position from a distance-velocity table as previously discussed with respect to the High Speed Mode of operation. However, in the Medium Speed Mode of operation, position of Transducing apparatus 120 is monitored with respect to every track crossing on Disk 122, i.e., with respect to every zero crossing of the sine signal.

Broadly stated, the Transition Mode of operation relates to the management of the positioning of Transducing apparatus 120 when it is less than the second preselected distance (3 tracks in the preferred embodiment), but greater than a third preselected distance (192 microinches in the preferred embodiment). In the Transition Mode of operation, the desired velocity of Transducing apparatus 120 is again determined by the remaining distance from the desired position from a distance-velocity table as previously discussed with respect to the High and Medium Speed Modes of operation. However, in the Transition Mode of operation, position information from Encoder 140 is more frequently monitored than was the case with the High and Medium Speed mode. In addition, both the sine and cosine signal from Encoder 140 are monitored. Monitoring of both the sine and cosine signals is necessary as both signals have non-linear regions therein, e.g, each signal is approximately linear within 45 degrees of each zero crossing, and becomes increasingly non-linear thereafter. Consequently, when one signal is in the non-linear region, position determination is accomplished by monitoring the alternate signal from Encoder 140, in a manner well known to those skilled in the art. As position information is known to a higher degree than was the case in the High and Medium Speed Modes, the velocity table used in the Transition Mode correspondingly provides desired velocity information in accordance with the higher degree of position accuracy.

Broadly stated, the Detent Mode of operation relates to the final positioning of Transducing apparatus 120 when Transducing apparatus 120 is less than the third preselected distance from the desired position (192 microinches in the preferred embodiment). The Detent Mode is a position control servo loop using a Position Integral Differentiation technique, well known to those skilled in the art, as more fully discussed in Control and Dynamic Systems by Takahasi, Rabins and Asulander, published by Addisson Wesley, November 1972, which is hereby incorporated by reference. As was the case with the Transition Mode, position information is again determined by the monitoring of both the sine and cosine signals from Encoder 140.

In accordance with the present invention, in the High Speed and Medium Speed Modes of operation, high resolution of the instantaneous position of Transducing apparatus 120 with respect to Disk 122 is not maintained. Rather, position information of Transducer apparatus 120 is only maintained with respect to the number of tracks which have been crossed, by updating Commutation Counter and Tracks To Go Counter in response to detected zero crossings of the sine signal from terminal 142 of Encoder 140. Velocity information of Transducer apparatus 120 is thereafter determined by measuring the time which elapse between successive track crossings.

In the preferred embodiment, external commands calling for a move to a desired new track are received on terminal 28 of Microprocessor 10, and consist of a first signal indicating the desired direction of motion, and a second signal comprising a series of pulses, the number of which indicating the number of tracks Transducing apparatus 120 is to be moved, in accordance with the Seagate ST506/ST412 Interface Standard. The number of pulses received are accumulated in the Tracks To Go Counter within Microprocessor 10.

When the contents of Tracks To Go Counter is not zero, a seek routine is initiated. FIGS. 5A-5E illustrate of the seek routine. Referring now to FIG. 5A, pulses indicating a desired amount of motion, as previously discussed, received on terminal 28 of Microprocessor 10 are first accumulated, as indicated in block 250. Motion is initiated whenever this count is not zero. The first step is to apply an acceleration current through Digital-to-Analog Converter 40 or 50 (FIG. 1), and the corresponding Motor Drive Amplifier 60 or 70 to motor windings 82 or 84, as indicated in block 252. The value of current which is supplied is determined based upon whether the destination track is greater than a preselected number of tracks away. In the preferred embodiment, the preselected number of tracks was 50. Consequently, a first maximum amount of current is supplied to a selected motor winding 82 or 84 if the number of selected tracks is greater than 50 tracks away, and a second maximum amount of current if the number of selected tracks is less than 50 but more than 3 tracks away. The maximum amount of current in each of the two cases is empirically determined from mechanical and electrical performance considerations unique to a particular design. As motion begins, the output of the cosine signal on terminal 144 of Encoder 140 is monitored in block 254 through Analog Multiplexer 90 and Analog-to-Digital Converter 110 until a zero crossing of the cosine signal is detected, as indicated by a change in polarity of the cosine signal. This confirms that motion has in fact begun. The zero crossing of the cosine signal is also used to update the Commutation Counter by one in block 256. Thereafter, Microprocessor 10 switches Analog Multiplexer 90 to monitor the sine signal from terminal 142 of Encoder 140, and in a similar manner determines the zero crossing of the sine signal in block 258, thereafter decrementing the Tracks To Go Counter by one in block 260. Microprocessor 10 thereafter in block 262 makes a comparison of the count in the Tracks To Go Counter with a second preselected distance. If the count is less than the preselected number of tracks, the Transition Mode of operation is thereafter entered, as will be more fully discussed hereinafter with respect to FIG. 5D. However, if the count is not less than the preselected number of tracks, the operational mode will remain in the High Speed or Medium Speed mode, as more fully discussed hereinafter. In the preferred embodiment, the preselected number of tracks is three.

Microprocessor 10 next determines in block 264 if the velocity of Transducing apparatus 120 is above a preselected velocity. If the velocity is above the preselected velocity, the High Speed Mode of operation is entered, and the operations indicated generally by block 266 (FIG. 5B) will next be performed, as more fully discussed hereinafter. If the velocity is less than the preselected velocity, the Medium Speed Mode of operation is entered, and the operations indicated generally by block 268 will thereafter be performed, as more fully discussed hereinafter. In the preferred embodiment, the preselected velocity is 6.7 inches per second.

At this time, as motion has just be initiated, it is known that the velocity of Transducing apparatus 120 cannot be above the preselected velocity. Initial positioning operations are therefor performed in the Medium Speed Mode. Microprocessor 10 will consequently next enter the Medium Speed Mode of operation by performing the operation broadly indicated in block 268 of FIG. 5B. Referring now to FIG. 5B, Microprocessor 10 updates Commutation Counter by one in block 268, with the update being either an increment or decrement depending upon the direction of motion of Transducer apparatus 120. In the Medium Speed Mode, position information of Transducing apparatus 120 is determined with respect to each zero crossing of the sine signal from terminal 142 of Encoder 140. Consequently, blocks 270, 272 and 274 relate to the monitoring of each crossing of the sine signal. Microprocessor 10 next determines whether the zero crossing of the sine signal was a change from a positive to a negative value, or a change from a negative value to a positive value, and on that basis continues to monitor the output from terminal 142 of Encoder 140 as previously discussed until the reverse transition at the next zero crossing. In particular, in block 270 Microprocessor 10 determines if the last zero crossing of the sine signal was a transition from a negative value to a positive value, and if so, next performs the operations indicated generally by block 274; otherwise performs the operations indicated in block 272. In block 274 Microprocessor 10 waits for the next transition of the sine signal from a negative value to a positive value, and thereafter performs the operation indicated generally in block 276. In block 272, Microprocessor 10 waits for the next transition of the sine signal from a positive value to a negative value, and thereafter performs the operation indicated generally in block 276. In block 276, Microprocessor 10 determines the time between the prior sine signal zero crossing and the time at which the immediately preceeding sine signal zero crossing occurred. This time difference is a direct indication of the velocity at which Transducing apparatus 120 is moving. Thereafter Microprocessor 10 decrements the Tracks To Go Counter by one in block 278. Microprocessor 10 next checks the value of the Tracks To Go Counter in block 280 to determine if Transducing apparatus 120 is at the second preselected distance (3 tracks in the preferred embodiment) from the desired destination. If Transducing apparatus 120 is the selected number of tracks from the desired destination, the Transition Mode of operation is entered, as will be more fully discussed hereinafter; otherwise, Microprocessor 10 next checks the velocity of Transducing apparatus 120 in block 282. In block 282, Microprocessor 10 compares the determined velocity of Transducing apparatus 120 with the previously discussed preselected velocity (6.7 inches per second in the preferred embodiment) to determine if Transducing apparatus 120 is moving at a greater velocity. If Transducing apparatus is moving at a greater velocity, operation in the High Speed Mode will be indicated by the setting of a Mode 2 Flag in block 286; otherwise, the Mode 2 Flag will be cleared in block 284 indicating the Medium Speed Mode in block 284.

Referring now to FIG. 5C, Microprocessor 10 next in block 288 uses the foregoing determined measured time between zero crossings of the sine wave to access a first look up table from which a corresponding velocity of Transducing apparatus 120 may be determined. From the velocity so determined from the first look up table, Microprocessor 10 next determines in block 290 if Transducing apparatus 120 is greater than the first preselected distance (108 tracks in the preferred embodiment) from the desired location. If Transducing apparatus 120 is greater than the first preselected distance, a preselected maximum velocity (10 inches per second in the preferred embodiment) is desired for Transducing apparatus 120 in block 292; if Transducing apparatus 120 is less than or equal to the first preselected distance, Microprocessor 10 thereafter performs the operations indicated generally by block 294.

In block 294, Microprocessor 10 checks the current value in the Tracks To Go Counter to determine the remaining distance to the desired destination track. The remaining distance is thereafter used with a second look up table to determine a corresponding desired velocity for Transducing apparatus 120 based upon the remaining distance. In this regard, the values in the second look up table follow the well known fractional power law for desired velocities of motion based upon remaining distance as previously discussed. Microprocessor 10 thereafter in block 296 determines the difference between the velocity previously determined in block 276 with the desired velocity determined from the second look up table in block 294 to determine a velocity error and thereafter multiplied by a gain constant. The velocity of Transducing apparatus 120 is thereafter adjusted in block 298 in accordance with the determined velocity error by a signal applied to the appropriate motor winding 82 or 84 through the associated Digital-to-Analog Converter 40 or 50 and the corresponding Motor Drive Amplifier 60 and 70.

Referring now to FIG. 5A, Microprocessor 10 next returns to block 264, and again determines if the velocity of Transducing apparatus 120 is greater than the preselected velocity (6.7 inches per second in the preferred embodiment). If the velocity is greater than the preselected velocity, operation in the High Speed mode is indicated.

In the High Speed Mode, the total computations to be done require more time for Microprocessor 10 to perform than is available between successive zero crossings of the sine wave. The control loop is therefore changed to measure every second zero crossing of the sine signal. Each such transition indicates that two tracks have been traversed since the last update of the several system counters. Therfore, in the High Speed Mode, the first step is to update the Commutation Counter by two tracks, which Microprocessor 10 performs in block 266 (FIG. 5B), and thereafter determines in block 300 if the last zero crossing of the sine signal was a transition from a negative value to a positive value, hereinafter referred to as a positive transition, or a transition from a positive value to a negative value, hereinafter referred to as a negative transition. If the previous transition was not a positive transition, Microprocessor 10 next in block 302 monitors the sine signal from terminal 142 of Encoder 140 and waits for a negative transition to occur, thereafter performing the operations indicated generally by block 304. If the previous transition was a positive transition, Microprocessor 10 waits for the next positive transition to occur in block 306. Microprocessor 10 determines the time at which the transition occurs in block 304, and determines therefrom the time from the preceeding transition. The value so determined is thereafter divided by two in block 304. Microprocessor 10 thereafter in block 306 decrements the Tracks To Go Counter by two, and thereafter performs the operations previously discussed with respect to block 280.

The maximum velocity in the High Speed Mode continues until the Tracks To Go Counter indicates that Transducer apparatus 120 is less than the preselected number of tracks from the desired destination (108 tracks in the preferred embodiment). When this occurs, the look up table provides the previously described reduced velocity, and subsequent velocity values progressively decrease as the Tracks To Go Counter decreases below the preselected number of tracks.

As the foregoing operation continues to control the velocity of Transducing apparatus 120, the Commutation Counter continues to determine which of the two coils of Motor 80 is the appropriate coil to be energized at any given time, and perform the necessary commutations as previously discussed.

The foregoing operation results in the velocity of Transducing apparatus 120 decreasing in a predetermined manner. When the measured velocity drops below the preselected velocity (6.7 inches per second in the preferred embodiment), the Mode 2 Flag is cleared in block 284, and the Medium Speed Mode of operation is entered.

The Medium Speed Mode continues to reduce the velocity of Transducing apparatus 120 until Microprocessor 10 determines that Transducing apparatus 120 is within 3 tracks from the desired destination, at which time the Transition Mode is entered.

When the Transition Mode is entered, the velocity of Transducing apparatus 120 has dropped to such a value that a measurement of velocity based on the time between zero crossings would not provide sufficient resolution to properly control Trasducing apparatus 120 in a desired manner. In this mode therefore, Microprocessor 10 begins to read both the sine and cosine signals of Encoder 140, i.e., the signals present on terminals 142 and 144 of Encoder 140, to read not only zero crossings, but instantaneous values. The instantaneous values of the sine and cosine provide continuous relative position information.

In the Transition Mode, Microprocessor 10 determines velocity of Transducing apparatus 120 on a regular basis by measuring the displacement difference at regular intervals. In the preferred embodiment, the time interval at which displacement is measured is 135 microseconds. The velocity so determined is utilized in the same manner as previously described in the High Speed and Medium Speed mode of operation, and compared with the desired velocity determined from the second look up table, in order to determine the velocity error, and thus control Motor 80 to achieve the desired velocity.

Microprocessor 10 determines the present distance from the desired destination using the Tracks To Go Counter and position information from the sine and cosine signals from Encoder 140. The position error thus derived is used in a third look up table to determine the desired velocity which is thereafter used to control excitation of Motor 80.

The sequence of steps associated with the Transition Mode begins with a positive determination in either block 262 (FIG. 5A) or block 280 (FIG. 5B), and are indicated in FIG. 5D. Referring now to FIG. 5D, Microprocessor 10 in block 308 up-dates the Commutation Counter by the number of tracks to go to the destination tracks (1 to 3 in the preferred embodiment). This is performed to ensure that if the Commutation Counter happens to be close to performing a commutation (within 3 tracks in the preferred embodiment), the commutation will be forced to take place at this time, rather at a later point. If the commutation were allowed to take place at a later point, undesirable transients could be introduced into the positioning of Transducing apparatus 120.

Microprocessor 10 next in block 310 gets current position information of Transducer apparatus 120 from the sine and cosine signals from Encoder 140. In block 312, Microprocessor 10 next determines the distance Transducer apparatus 120 is from the desired destination using the sine and cosine values and the information in Tracks To Go Counter. In block 314, Microprocessor next determines if the distance to go is less than the third preselected distance (192 microinches in the preferred embodiment). If the distance is less than the third preselected distance, the Detent Mode of operation is entered at block 326 in FIG. 5E, otherwise block 316 is next performed. In block 316, Microprocessor 10 determines the desired velocity for Transducing apparatus 120 from the previously discussed look up table, and in block 318 determines the current velocity of Transducing apparatus 120. In block 320 Microprocessor 10 next determines the difference between the desired velocity and the measured velocity of Transducing apparatus 120, and in block 322 adjusts the value so obtained by a velocity gain error. The velocity gain error is a constant which is empirically determined by parameters associated with a particular design. In block 324, Microprocessor 10 outputs the value so determined to the appropriate digital-to-analog converter to drive the motor winding. Thereafter, Microprocessor 10 next repeats the foregoing described procedure starting with the operations previously discussed with respect to block 310.

When the position error decreases to a value less than a preselected distance, the Detent Mode of operation is entered, i.e., with a positive determination in block 314 (FIG. 5D). In the preferred embodiment, the preselected distance is 192 microinches.

In the Detent Mode of operation, velocity is determined in a similar manner as in the Transition Mode, except that instead of making a comparison every 135 microseconds, the comparison of position is made every 70 microseconds. Position error is determined in the same way as in the Transition Mode previously discussed, but requires reference only to the signal being derived from the sine signal of Encoder 140. This is possible due to the fact that as Transducing apparatus 120 is within 192 microinches of a zero crossing, corresponding position information for the sine signal is on the linear portion thereof, i.e., close to a sine signal zero crossing. Consequently, position information can therefore be determined exclusively from this signal. Microprocessor 10 must, however, determine whether the desired track center location is defined by the positive going or negative going slope of the sine signal, which is done by reference to the Track Crossing Counter. In the Detent Mode, control of position is accomplished by employing the Position Integral Differentiation approach, as previously referenced. Using the Position Integral Differentiation approach, a sum is determined of velocity error, position error and the integral of the position signal. Velocity and position error have been previously described. The position integral is developed by Microprocessor 10 by adding the position error determined to a register which builds up the integral of the position error by successive additions of the position error. The previously described servo burst correction offset is further added to the position signal derived from the sine signal from Encoder 140, as discussed hereinafter.

The operations performed in the Detent Mode are illustrated in FIG. 5E. Referring now to FIG. 5E, Microprocessor 10 determines which slope of the sine wave Transducing apparatus 120 is positioned with respect to, and thereafter initiates the integrator in block 326. Thereafter in block 328 position information is determined from the sine signal, and thereafter converted to actual position information in block 330. As was previously discussed, in the Detent Mode of operation, servo burst information is also used with respect to positioning of Transducer apparatus 120. This is necessary to provide position compensation for thermal expansion which may occur with respect to disk 122 (FIG. 1), as previously discussed. Consequently, in block 332 position correction information determined from servo burst detection and processing as previously discussed is added to the position error determined. The position so determined is thereafter multiplied by a gain factor which is empirically determined by parameters associated with a particular design. Thereafter in block 334 the velocity error is determined, and again multiplied by a gain factor as previously discussed. Thereafter in block 336, the present position error, previous position error, and previous integrator value are added to get a new integral value, which is likewise multiplied by a gain factor as previously discussed. Thereafter in block 338, the position error, velocity error and integral error are added, and thereafter coupled to the appropriate digital-to-analog converter in block 340. Thereafter, the preceeding operations are again repeated beginning with block 328.

While in the Detent Mode, servo burst information is monitored. Once each revolution of Disk 122, as indicated by the presence of the Index pulse from Transducer 126, Microprocessor 10 controls Gated Integrator 130 to measure the values of servo bursts 202 and 204 (FIG. 3) on the track over which Transducing apparatus 120 is positioned. The magnitudes of the respective servo bursts are compared and used with previously determined servo burst information to determine servo burst correction. The operations performed by Microprocessor 10 with respect to the processing of the servo burst information are illustrated in FIG. 5E beginning with block 342. In block 342, responsive to the occurrence of an index pulse occurring on terminal 26 of Microprocessor 10 (FIG. 1), Microprocessor 10 is interrupted from its current operation in block 344. Thereafter in block 346 the amplitude of the first servo burst is read by Microprocessor 10, i.e. servo burst 202 (FIG. 3). Thereafter in block 348, the amplitude of the following servo burst is read by Microprocessor 10, i.e., servo burst 204. Microprocessor 10 next determines the difference between the amplitudes of each of the servo bursts in block 350. In the preferred embodiment, only one bit of the resulting difference is used for correction purposes on each reading of the servo bursts. This is done to guard against invalid information corrupting the positioning process. Consequently in block 352, the value determined from block 350 is adjusted to reflect only the least significant bit, which is thereafter added to the servo offset value in block 354. In block 356 the sign is changed as determined by the slope of the sine signal as previously discussed. Thereafter, Microprocessor 10 next performs the sequence of steps previously discussed, starting with block 332.

While the foregoing has provided a detailed description of the operation of the transducer positioning apparatus in accordance with the present invention, yet further detailed information regarding the foregoing described operation is included with the commented assembly language listing of the previously discussed program of Microprocessor 10 in Table 1 hereinbelow. ##SPC1##

It is consequently observed from the foregoing that an improved method and apparatus has been described which provides a number of significant advances in the art. In particular, increased range of motion of a transducer apparatus is achieved through the commutation of a prime mover associated with the transducing apparatus. In addition thereto, an improved method of control has been disclosed employing digital techniques. While the foregoing has been described in terms of a particular preferred embodiment, it should be understood that such description is not to be interpreted in a limiting manner. To the contrary, the present invention is only to be limited by the scope of the following claims.